A Methodology Study of Integrating Systems Thinking and Bayesian Network in Health Policy

Project: A - Government Institutionb - National Science and Technology Council

Project Details


Health policy research was given much less attention comparing to other researches of health care services, especially in the study of its methodology. Owing to the system complexity and uncertainties of events in health policy, in 2012, a methodology study of integrating systems thinking (system dynamics) and Bayesian network in health policy was proposed and approved. This project applies the long-term care system of patients with prolonged mechanical ventilator in Taiwan as a targeted health policy in order to develop a valid and effective method for the foundation of health policy research in the future. This proposed project is to extend the previous study of ‘A methodology study of integrating systems thinking and Bayesian network in health policy (I)’. The research will be conducted based on a specifically designed three-year project blue print. It mainly follows the accomplished or will be accomplished tasks by the end of the first year study, which includes literature review, summary of policy changes of long-term respiratory care system, the closed feedback diagram of long-term respiratory care system focusing on the hospital management subsystems only, and a preliminary framework of integration of systems thinking and Bayesian network. The future two-year project will extend the system boundary to the six regional administration systems of national health insurance, implementation of integrating systems thinking and Bayesian network, involvement of population subsystem, simulation analysis and evaluation based on different policy sensoria, and summarization of the methodology knowledge of health policy research by year accordingly.
Effective start/end date8/1/1411/30/15


  • health policy
  • systems thinking
  • Bayesian network
  • mixed methods
  • methodology


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.